Morphological Brain Age Prediction using Multi-View Brain Networks Derived from Cortical Morphology in Healthy and...

Area of Science:

• Neuroscience
• Radiology
• Computational Biology

Background:

• Brain development and aging are complex processes studied using neuroimaging.
• Discrepancies between chronological and data-driven brain age are observed using functional MRI (fMRI) and diffusion MRI (dMRI).
• Predicting brain age from connectomic data can reveal biomarkers for neurological disorders.

Purpose of the Study:

• To predict brain's morphological age using solely morphological brain networks derived from T1-weighted MRI.
• To identify connectional brain features related to age in autistic and healthy populations.
• To outperform existing brain-age prediction methods by utilizing a connectomic perspective.

Main Methods:

• Building multi-view morphological brain networks (M-MBN) from T1-weighted MRI data.
• Performing feature extraction and selection from M-MBN.
• Training a machine learning regression model to predict age from M-MBN data.
• Applying the model to identify age-related connectional brain features in autistic and healthy individuals.

Main Results:

• The proposed method significantly outperforms existing brain-age prediction approaches.
• Discovered connectional brain morphological features that accurately fingerprint brain age in both autistic and healthy individuals.
• Identified that connectional cortical thickness is the strongest predictor of morphological age in the autistic brain.

Conclusions:

• Morphological brain networks derived from T1-weighted MRI can accurately predict brain age.
• This novel connectomic approach offers a powerful tool for understanding brain aging and neurological disorders.
• Connectional cortical thickness is a crucial feature for assessing morphological brain age in autism.